The printer worked well (no duplexing jams) for the first couple of years I had it. This is new behavior from the same old firmware.

Tray 2 (the problem tray) sits at the bottom of the printer, slides out like a dresser drawer to replenish paper. The paper is picked up from tray 2, goes up the back side of the printer around the print drum, then through the fuser to fix the toner, then if duplexing is set it goes most of the way into the output tray then gets sucked back through the duplexer and around the drum to print the back side of the page, through the fuser again, and finishes in the output tray.

That's the puzzling part. The paper is only fed to the duplexer at the end of the process, after it has gone through the drum and the fuser, long after it has left the tray. Oh, wait...

The first page is picked up, goes through drum, fuser, duplexer, back to the drum and fuser for the second side, and the next page is being picked up before this process is finished -- probably about the time the second side of the first page is going through the fuser. This is true for non-duplex printing as well, but with duplexing there is the delay while the second side is being printed before the next page is picked up. It is different logic for the paper handling, not to mention a different path.

I don't think that's it, because if the paper makes it out of the tray and onto the print drum, the duplexer itself is 100% reliable. Tray-->Drum-->Fuser-->Duplexer-->Drum-->Fuser-->Output Tray.

Tray 1 tilts out from the front of the printer (see photo) and is much lower capacity, meant for single sheet feeding, envelopes, card stock (the turns in the paper path are less acute), etc. From the time the paper starts up the back of the printer towards the print drum, the paper path is the same; it's only the initial pickup that differs.

I think you have put your finger on it. When the different logic is activated for duplexing, the pick roller is picking the paper out of the tray, but the feed/registration rollers are not activating to pull the paper into the printer during the time in which the previous page is still wending its way through the fuser and into the output tray.

Unless this is a known problem that other people have figured out how to fix, I don't think there's much I can do about it. At least I still have tray 1.

tanstaafl.